75% of all breast tumors in women express estrogen receptor alpha (ER?+). Although anti-estrogens have markedly reduced mortality for many women with this luminal tumor subtype, one quarter of these patients eventually succumb to anti-estrogen-resistant cancer. Tumor initiating cells, or cancer stem cells have been implicated in resistance to conventional therapies, recurrence and metastasis. However, the origin of luminal tumors and the characteristics of the cells that defy treatment remain unknown. Mouse models of ER?+ breast cancer that would facilitate these studies are very rare. Multiple studies point to a close link between prolactin (PRL), aand the pathogenesis and therapeutic responsiveness of ER?+ breast cancer. Our in vivo model, the NRL-PRL transgenic mouse, permits dissection of the dynamic processes that lead to diverse carcinomas with transcript profiles that resemble the luminal tumor subtype in women, including ER? expression. In this new application, we will employ this model to investigate the hypothesis that PRL cooperates with ovarian steroids to modulate mammary epithelial progenitor and tumor cell subpopulations, promoting the development of luminal carcinomas. Further, these tumors contain tumor initiating cells, which display hormonal dependencies and therapeutic susceptibilities distinct from the bulk of tumor cells. In Aim 1, we will establish the effect of PRL and crosstalk with ovarian steroids on mammary epithelial subpopulations in nonparous females prior to lesions, as defined by surface markers and functional progenitor assays, and identifies key mediators of these processes. In Aim 2, we will determine characteristics of tumor initiating cells in PRL-induced ER?+ carcinomas, and ascertain their dependence on PRL and ovarian steroids and susceptibility to adjuvant treatment. In Aim 3, we will elucidate interplay between PRL and a well-characterized oncogenic signal, ?-catenin, in regulation of normal mammary subpopulations and tumor subtype by examining the net effect of crosstalk on epithelial subpopulations prior to lesions, in early lesions, and carcinomas. These studies will illuminate the mechanism(s) whereby PRL interacts with endogenous steroids and oncogenes to promote the luminal breast cancer subtype, and its role in treatment sensitivity. They will test the cancer stem cell hypothesis in PRL-induced ER?+ carcinomas, generating a new and biologically relevant paradigm for understanding this prevalent tumor subtype, enabling the design of novel therapeutic approaches.